Six 2026 predictions from AT&T about the thing we love and loathe at the same time

How companies should in 2026 – and beyond.

0comments
AT&T logo in blue.
AT&T has something to share with us about AI.

No, no, no – Ma Bell is not announcing that it's going to be the first company to stay away from AI. Instead, AT&T has some AI-related predictions about 2026 that are worth checking out.

2026 is almost here


The telco says all companies need to brace for the significant changes AI will trigger throughout 2026.

AT&T Chief Data Officer Andy Markus lays it all out and notes that in order to stay competitive, a company should stay ahead of the progressions. In his words, the AI (r)evolution will happen with the help of AI agents and AI-fueled coding.

1. Fine-tuned SLMs for the win


Enterprises will mostly rely on fine-tuned small language models (SLMs) rather than giant, general-purpose ones. These smaller models are trained on a company's own data for specific tasks, making them accurate, fast, and cheaper to run. Large models (LLMs) will still coordinate complex workflows, but SLMs will do most of the day-to-day work inside those systems. As businesses mature in their AI use, they will see that fine-tuning smaller models is the best way to turn internal data into real value.

2. AI-coding: the new standard


AI-driven coding will become a standard way to build software, cutting some development cycles down to minutes. AI already helps with basic coding tasks, but it will soon handle much larger parts of the process. Developers will move faster, focus more on design and problem-solving, and switch roles more easily across a project. In some cases, entire apps will be built almost in one pass, with minimal human edits. Even non-technical teams will be able to create software using plain language, with AI turning those ideas into production-ready code.

3. Apps on demand


Companies will start building apps on demand, powered by AI agents. Instead of investing years into traditional software, businesses will create apps quickly when they are needed and update them just as fast. AI agents will be able to adjust to new requirements without starting from scratch. Traditional apps will not disappear right away, but on-demand AI-built apps will often be a faster and cheaper option for solving immediate business problems.

Recommended For You

4. Private high-speed fiber networks


Large enterprises will use private, high-speed fiber networks to link directly to centralized computing centers. As AI workloads grow, fast and reliable connectivity will become just as important as computing power. Companies investing heavily in AI will build dedicated fiber connections to their cloud and compute resources. Over time, cloud providers may move data centers closer to these private networks, blurring the lines between cloud, hybrid, and on-prem systems.

5. Telcos will go for productized AI offerings


Telecom companies will expand beyond connectivity and offer more AI services to businesses. With their experience handling massive amounts of data and training models, telcos are well positioned to provide services like model fine-tuning. While not announcing new AT&T products, the prediction is that market demand and technical expertise will push telecom firms into productized AI offerings. This would create new revenue streams and deepen their role in AI-driven business operations.

6. The AI performance


Businesses will closely track AI performance using clear metrics like accuracy, speed, and cost. As AI becomes essential across all departments, companies will focus less on experimenting and more on measuring results. Teams will evaluate return on investment, reliability, and scalability for every AI tool they use. Over time, these metrics will become a shared way for organizations to talk about and manage AI across the entire business.

Are you ready for the feature?



AI is what gets people polarized big time right now – and I imagine that it'll continue to do so in the future (not just in 2026, but for many years to come).

I agree that big companies and corporations should definitely look into using AI, but with a measure. It's easy to get overhyped, overinvest and burn yourself. To end on a positive note, I still can't forget that story about how it was precisely AI that turned out to be slowing down senior coders.

So, yeah: take everything with a pinch of salt!

Do you see yourself relying more and more on AI in 2026?

Google News Follow
Follow us on Google News
COMMENTS (0)

Latest Discussions

by 30zpark • 2
by RxCourier9534 • 13

Recommended For You

FCC OKs Cingular\'s purchase of AT&T Wireless